l82 DAVENPORT 



yield only the dominant type and the recessive type respec 

 lively. These expectations have been now so often realized 

 that Mendel's law has gained deserved fame as the most im- 

 portant law of inheritance yet enunciated. 



Many illustrations of this law are given by poultry. Thus 

 when frizzled fowl are crossed with non-frizzled, the offspring 

 are frizzled ; from such frizzled hybrids 25 per cent, plain- 

 feathered are derived. These plain-feathered birds have not, 

 in 50 offspring, thrown one frizzled fowl. 



When the silky fowl is crossed with plain-feathered fowl, all 

 of the offspring have plain feathers. When the hybrids are 

 mated 25 per cent, silky offspring result. Such silky offspring 

 breed true. And so I might continue with a dozen illustrations 

 drawn from my experience demonstrating the truth of Mendel's 

 law. 



As the truth of a new law becomes apparent, enthusiasm for 

 it is apt to carry one to extremes — to declare its universality 

 and to overlook exceptions. It is well to cultivate the skeptical 

 frame of mind and to remember that what looks like the whole 

 truth may be only partial and that an examination of the ex- 

 ceptions may lead to larger generalities. 



First of all we find that even in cases of Mendelian inheri- 

 tance in the second hybrid generation, the recessives that are 

 supposed to come from two pure recessive gametes show in 

 their somas traces of the dominant type. Thus when a crested 

 bird is crossed with a plain-headed one, and the crested hybrids 

 are then crossed inter se the extracted recessives of the second 

 hybrid generation are plain-headed to be sure, but they show a 

 disturbance of certain feathers. Again foot-feathering, or 

 booting, is dominant, but in the cross between the Dark* Brahma 

 and Minorca small feathers can frequently be found on the feet 

 of all the second hybrid generation where 25 per cent, bootless 

 is expected. The boot character is not absent from the 25 per 

 cent., but it is immensely reduced in expression as compared 

 with the booted ancestor. 



Second, we find that the two opposed unit characters are not 

 always respectivel}'^ dominant and recessive. Various classes 

 can be recognized. In some cases both characters appear in tlie 



